Sunteți pe pagina 1din 3

Downloaded from on August 2, 2016.


See the related article beginning on page 507. increased hexosamine pathway flux
(1, 2, 10, 11). While specific inhibitors
Oxidative stress and diabetic neuropathy: of each pathway block one or more
diabetic microvascular complications,
a new understanding of an old problem only recently has a link been estab-
lished that provides a unified mecha-
nism of tissue damage. Each pathway
Eva L. Feldman becomes perturbed as a direct or indi-
rect consequence of hyperglycemia-
Juvenile Diabetes Research Foundation Center for the Study of Complications in Diabetes, and
the Department of Neurology, University of Michigan, Ann Arbor, Michigan, USA
mediated superoxide overproduction
by the mitochondrial electron trans-
J. Clin. Invest. 111:431433 (2003). doi:10.1172/JCI200317863. port chain. Either inhibition of super-
oxide accumulation or euglycemia
restores the metabolic and vascular
Diabetes has reached epidemic pro- diabetes. Diabetic retinopathy is the imbalance and blocks both the initia-
portions in the Western world. In the most common cause of adult blind- tion and progression of complica-
United States, 17 million individuals ness in the United States. Ninety per- tions (2, 10, 12).
have diabetes, greater than 6% of the cent of diabetics present evidence of In the diabetic state, unchecked
population (1). The morbidity and retinopathy within 15 years of disease superoxide accumulation and result-
mortality of diabetes is due to the onset and approximately 25,000 new ant increases in polyol pathway activ-
development of both macrovascular cases of diabetes-related blindness are ity, AGE accumulation, PKC activity,
and microvascular complications (2). reported per year (5). Diabetes is also and hexosamine flux trigger a feed-
Macrovascular complications includ- the leading cause of renal failure in the forward system of progressive cellu-
ing myocardial infarction, stroke, and United States, accounting for 40% of lar dysfunction (Figure 1). In nerve,
large vessel peripheral vascular disease new cases each year (6). Greater than this confluence of metabolic and vas-
are 2 to 4 times more prevalent in indi- half of all patients with diabetes devel- cular disturbances leads to impaired
viduals with diabetes. The underlying op neuropathy, a progressive deterio- neural function and loss of neu-
common factor in macrovascular com- ration of nerves resulting in peripher- rotrophic support, and long term,
plications is the ability of the diabetic al and autonomic nerve dysfunction. can mediate apoptosis of neurons
condition to accelerate atherogenesis. As a result, diabetic neuropathy is the and Schwann cells, the glial cells of
Atherogenesis is a multifactorial most common cause of nontraumatic the peripheral nervous system
response of vessels to injury; both amputations and autonomic failure (1315). Decreases in nerve growth
insulin resistance and elevated lipid (7, 8). In his or her lifetime, a diabetic factor (NGF), neurotrophin-3 (NT-3),
levels, common in diabetes, are pri- patient with neuropathy has a 15% ciliary neurotrophic factor, and
mary triggers of atherogenic injury (3). chance of undergoing one or more IGF-I in nerves from animals with
The endothelium in diabetic arteries is amputations (9). experimental diabetes are well docu-
also more prone to atherogenic injury, mented and correlate with the pres-
likely due to decreased production of What are the mechanisms that ence of neuropathy (1618).
endothelial nitric oxide, known to be underlie the development of
antiatherogenic, and increased pro- microvascular complications? Hedgehog proteins and
duction of plasminogen activator Similar to our understanding of diabetic neuropathy
inhibitor-1 (PAI-1) (4). While macro- macrovascular complications, it is The elegant work of Calcutt and col-
vascular complications are common becoming increasingly clear that leagues in this issue of the JCI reports
among diabetics, diabetes-specific microvascular complications share a a decrease in desert hedgehog expres-
microvascular complications will even- common pathophysiology. Animal sion in nerves from young adult rats
tually affect nearly all individuals with and in vitro experiments over the last with streptozotocin-induced diabetes
25 years have implicated four major (19). Hedgehog proteins (sonic, desert,
Address correspondence to: Eva L. Feldman, pathways of glucose metabolism in and indian) are essential for normal
JDRF Center for the Study of Complications the development of microvascular nervous system development (20).
in Diabetes, and the Department of complications (10). These include: 1) Desert hedgehog is found exclusively
Neurology, University of Michigan, increased polyol pathway activity in the peripheral nervous system in
4414 Kresge III, 200 Zina Pitcher Place,
Ann Arbor, Michigan 48109, USA. leading to sorbitol and fructose accu- Schwann cells and is important in
Phone: (734) 763-7274; Fax: (734) 763-7275; mulation, NAD(P)H-redox imbal- peripheral nerve patterning (20). Af-
E-mail: ances, and changes in signal trans- ter 10 weeks of experimental diabe-
Conflict of interest: The author has declared duction; 2) nonenzymatic glycation tes, Calcutt et al. observed a decrease
that no conflict of interest exists. of proteins yielding advanced glyca- in desert hedgehog gene expression.
Nonstandard abbreviations used:
plasminogen activator inhibitor-1 (PAI-1);
tion end-products (AGEs); 3) activa- This decrease correlates with several
advanced glycation end-product (AGE); nerve tion of PKC thereby initiating a cas- well established physiological and
growth factor (NGF); neurotrophin-3 (NT-3). cade of stress responses, and 4) biochemical markers of experimental

The Journal of Clinical Investigation | February 2003 | Volume 111 | Number 4 1

Downloaded from on August 2, 2016.

Figure 1
Mechanisms leading to neuronal degeneration in hyperglycemia involve reactive oxygen species (ROS) formation. The diabetic state produces impaired
neurotropism, axonal transport and gene expression through at least four major pathways. 1) Excess glucose is diverted away from glycolysis by the
polyol pathway that depletes NADPH and cellular antioxidant capacity. 2) Glucose also may become oxidized and form AGEs that alter extracellu-
lar matrix, activate receptors that produce ROS intermediates, and alter intracellular protein function. 3) PKC becomes activated either directly by
glycolytic intermediates or indirectly as shown as a second messenger for stress hormones, leading to increased vascular disease, inflammation, and
oxidative stress. 4) Partial glycolysis causes accumulation of glycolytic intermediates and leads to escape of fructose-6-phosphate along the hex-
osamine pathway that increases vascular disease and further ROS generation. These mechanisms are ultimately linked to superoxide production
through increased glucose respiration that produces superoxide in the mitochondria and also activates the superoxide-producing NADH oxidase.
GSSG, glutathione disulfide; TCA, tricarboxylic acid cycle.

diabetes, including slowed motor and had a decrease in medium sized myeli- parameters to normal. Administration
sensory nerve conduction velocities, nated fibers, which was restored by of NGF restores neuropeptide levels
decreased nerve blood flow, decreased sonic hedgehog treatment. and sensory amplitudes in experimen-
pain threshold in response to heat tal diabetes (17, 21); in parallel, NT-3
and/or formalin, and decreased NGF Combination therapy for the normalizes nerve conduction slowing
and neuropeptide levels. Thrice week- treatment of diabetic neuropathy (22, 23) and IGF-I administration
ly injections of sonic hedgehog linked While purely speculative, it is likely blocks the development of neuropathy
to an IgG fusion protein, beginning that restoration of hedgehog activity and reverses impaired nerve regenera-
after 5 weeks of experimental diabetes provided much needed neurotrophic tion (24, 25). When oxidative stress is
and continuing for an additional 5 support both directly, by activat- induced in nerves of nondiabetic ani-
weeks, restored motor and sensory ing hedgehog downstream pathways mals by administering pro-oxidants,
nerve conduction velocities and both and indirectly, by restoring NGF levels. decreases in NGF and NT-3 are
NGF and neuropeptide levels. There As discussed above, hyperglycemia- observed similar to those reported in
was no therapeutic effect on nerve induced decreases in neurotrophic fac- animals with experimental diabetes
blood flow or pain threshold levels. tors are well documented, with neu- (26). Antioxidant therapy in experi-
Morphometric analyses of sciatic rotrophic replacement frequently mental diabetic neuropathy blocks the
nerves revealed that diabetic animals restoring one or more impaired nerve observed decreases in nerve NGF and

2 The Journal of Clinical Investigation | February 2003 | Volume 111 | Number 4

Downloaded from on August 2, 2016.

restores nerve function (27). Antioxi- 1. Windebank, A.J., and Feldman, E.L. 2001. Diabetes factors and peripheral neuropathy. In Diabetic
and the nervous system. In Neurology and general neuropathy. P.J. Dyck and P.K. Thomas, editors.
dant therapy also restores normal medicine. M.J. Aminoff, editor. Churchill Living- W.B. Saunders Co. Philadelphia, Pennsylvania,
blood flow and nerve conduction stone. Philadelphia, Pennsylvania, USA. 341364. USA. 377386.
velocities in experimental diabetes (11, 2. Brownlee, M. 2001. Biochemistry and molecular 19. Calcutt, N.A., et al. 2003. Therapeutic efficacy of
cell biology of diabetic complications. Nature. sonic hedgehog protein in experimental diabet-
12, 2832). Interestingly, neurotroph- 414:813820. ic neuropathy. J. Clin. Invest. 111:507514.
ic factors may also serve as antioxi- 3. Assmann, G., et al. 1999. Coronary heart disease: doi:10.1172/JCI200315792.
dants and this function may con- reducing the risk: a worldwide view. Internation- 20. Parmantier, E., et al. 1999. Schwann cell-derived
al Task Force for the Prevention of Coronary Desert hedgehog controls the development of
tribute to their role as possible Heart Disease. Circulation. 100:19301938. peripheral nerve sheaths. Neuron. 23:713724.
therapeutic entities in diabetic neu- 4. Cullen, P., von Eckardstein, A., Souris, S., Schulte, 21. Goss, J.R., et al. 2002. Herpes simplex-mediated
ropathy (3336). H., and Assmann, G. 1999. Dyslipidaemia and gene transfer of nerve growth factor protects
cardiovascular risk in diabetes. Diabetes Obes. against peripheral neuropathy in streptozotocin-
Currently, there are no treatments for Metab. 1:189198. induced diabetes in the mouse. Diabetes.
neuropathy other than treating the dia- 5. Gardner, T.W., Antonetti, D.A., Barber, A.J., 51:22272232.
betic condition per se. Our improved LaNoue, K.F., and Levison, S.W. 2002. Diabetic 22. Mizisin, A.P., Calcutt, N.A., Tomlinson, D.R., Gal-
retinopathy. More than meets the eye. Surv. Oph- lagher, A., and Fernyhough, P. 1999. Neu-
understanding of the pathogenesis of thalmol. 47(Suppl. 2):S253S262. rotrophin-3 reverses nerve conduction velocity
diabetic neuropathy should assist in 6. Skyler, J.S. 2001. Microvascular complications. deficits in streptozotocin-diabetic rats. J. Peripher.
the development of new therapies (7, 8, Retinopathy and nephropathy. Endocrinol. Metab. Nerv. Syst. 4:211221.
10). Clearly, therapeutic efficacy in man Clin. North Am. 30:833856. 23. Pradat, P.F., et al. 2001. Continuous delivery of
7. Feldman, E.L., Stevens, M.J., Russell, J.W., and neurotrophin 3 by gene therapy has a neuropro-
will be more challenging than in ani- Greene, D.A. 2002. Somatosensory neuropathy. tective effect in experimental models of diabetic
mal models of experimental diabetes. In Ellenberg and Rifkins diabetes mellitus. D. Porte, and acrylamide neuropathies. Hum. Gene Ther.
Damage to the nervous system by dia- Jr., R.S. Sherwin, and A. Baron, editors. McGraw 12:22372249.
Hill. New York, New York, USA. 771788. 24. Zhuang, H.X., Synder, C.K., Pu, S.F., and Ishii,
betes in man is more chronic, extensive, 8. Vinik, A.I., et al. 2002. Diabetic autonomic neu- D.N. 1996. Insulin-like growth factors reverse or
and severe (7, 8, 10). Several clinical tri- ropathy. In Ellenberg and Rifkins diabetes mellitus. arrest diabetic neuropathy: effects on hyperalge-
als have already failed to show improve- D. Porte, Jr., R.S. Sherwin, and A. Baron, editors. sia and impaired nerve regeneration in rats.
McGraw Hill. New York, New York, USA. Exp. Neurol. 140:198205.
ment of diabetic neuropathy in 789804. 25. Schmidt, R.E., et al. 1999. Insulin-like growth fac-
patients with type 1 and 2 diabetes 9. Feldman, E.L., Stevens, M.J., Russell, J.W., and tor I reverses experimental diabetic autonomic
(reviewed in ref. 1). For example, there Greene, D.A. 2001. Diabetic neuropathy. In Prin- neuropathy. Am. J. Pathol. 155:16511660.
ciples and practice of endocrinology and metabolism. 26. Hounsom, L., Corder, R., Patel, J., and Tomlinson,
is no therapeutic benefit of acetyl-car- K.L. Becker, editor. Lippincott Williams & D.R. 2001. Oxidative stress participates in the
nitine, aldose reductase inhibitors, or Wilkins. Philadelphia, Pennsylvania, USA. breakdown of neuronal phenotype in experimen-
NGF in human diabetic neuropathy 13911399. tal diabetic neuropathy. Diabetologia. 44:424428.
10. Stevens, M.J., Obrosova, I., Pop-Busui, R., Greene, 27. Obrosova, I.G., Fathallah, L., and Stevens, M.J.
(1). What is required is combination D.A., and Feldman, E.L. 2002. Pathogenesis of 2001. Taurine counteracts oxidative stress and
therapy. Similar to how the oncologist diabetic neuropathy. In Ellenberg and Rifkins dia- nerve growth factor deficit in early experimental
approaches cancer, the endocrinologist betes mellitus. D. Porte, Jr., R.S. Sherwin, and A. diabetic neuropathy. Exp. Neurol. 172:211219.
Baron, editors. McGraw Hill. New York, New 28. Low, P.A., Nickander, K.K., and Tritschler, H.J.
and/or neurologist could, in the future, York, USA. 747770. 1997. The roles of oxidative stress and antioxi-
approach diabetic neuropathy. By 11. Cameron, N.E., Eaton, S.E., Cotter, M.A., and Tes- dant treatment in experimental diabetic neu-
blocking multiple pathway compo- faye, S. 2001. Vascular factors and metabolic ropathy. Diabetes. 46(Suppl. 2):S38S42.
interactions in the pathogenesis of diabetic neu- 29. Tomlinson, D.R. 1998. Future prevention and
nents (Figure 1), multiple causes of ropathy. Diabetologia. 44:19731988. treatment of diabetic neuropathy. Diabetes Metab.
oxidative stress would in turn be 12. Greene, D.A., Obrosova, I., Stevens, M.J., and 24(Suppl. 3):7983.
blocked, preventing nervous system Feldman, E.L. 2000. Pathways of glucose-mediat- 30. Cameron, N.E., and Cotter, M.A. 1997. Metabol-
ed oxidative stress in diabetic neuropathy. In ic and vascular factors in the pathogenesis of dia-
injury. If coupled with additional Antioxidants in diabetes management. L. Packer, P. betic neuropathy. Diabetes. 46(Suppl. 2):S31S37.
antioxidant therapy or neurotrophic Rosen, H.J. Tritschler, G.L. King, and A. Azzi, edi- 31. van Dam, P.S., and Bravenboer, B. 1997. Oxidative
support, this type of complication tors. Marcel Dekker Inc. New York, New York, stress and antioxidant treatment in diabetic neu-
USA. 111119. ropathy. Neurosci. Res. Commun. 21:4148.
cocktail could provide the first effec- 13. Russell, J.W., Sullivan, K.A., Windebank, A.J., Her- 32. Greene, D.A., Stevens, M.J., Obrosova, I., and
tive treatment for diabetic neuropathy. rmann, D.N., and Feldman, E.L. 1999. Neurons Feldman, E.L. 1999. Glucose-induced oxidative
undergo apoptosis in animal and cell culture stress and programmed cell death in diabetic
models of diabetes. Neurobiol. Dis. 6:347363. neuropathy. Eur. J. Pharmacol. 375:217223.
Acknowledgments 14. Russell, J.W., et al. 2002. High glucose-induced 33. Pan, Z., Sampath, D., Jackson, G., Werrbach-
The author thanks Judy Boldt for oxidative stress and mitochondrial dysfunction Perez, K., and Perez-Polo, R. 1997. Nerve growth
excellent secretarial assistance and in neurons. FASEB J. 16:17381748. factor and oxidative stress in the nervous system.
Andrea Vincent for assistance with the 15. Schmeichel, A.M., Schmelzer, J.D., and Low, P.A. Adv. Exp. Med. Biol. 429:173193.
2003. Oxidative injury and apoptosis of dorsal 34. Park, D.S., et al. 1998. Multiple pathways of neu-
figure. This work was supported by root ganglion neurons in chronic experimental ronal death induced by DNA-damaging agents,
grants from the National Institutes of diabetic neuropathy. Diabetes. 52:165171. NGF deprivation, and oxidative stress. J. Neurosci.
Health (NS38849 and NS36778), by 16. Apfel, S.C. 1999. Neurotrophic factors and dia- 18:830840.
betic peripheral neuropathy. Eur. Neurol. 35. Delanty, N., and Dichter, M.A. 1998. Oxidative
the Juvenile Diabetes Research Foun- 41(Suppl. 1):2734. injury in the nervous system. Acta Neurol. Scand.
dation Center for the Study of Com- 17. Tomlinson, D.R., Fernyhough, P., and Diemel, 98:145153.
plications in Diabetes, and by the Pro- L.T. 1997. Role of neurotrophins in diabetic neu- 36. Lieberthal, W., Triaca, V., Koh, J.S., Pagano, P.J.,
ropathy and treatment with nerve growth factors. and Levine, J.S. 1998. Role of superoxide in apop-
gram for Understanding Neurological Diabetes. 46:S43S49. tosis induced by growth factor withdrawal. Am. J.
Diseases (PFUND). 18. Feldman, E.L., and Windebank, A.J. 1998. Growth Physiol. 275:F691F702.

The Journal of Clinical Investigation | February 2003 | Volume 111 | Number 4 3